{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:R2NF2DKDJJUR6OO2YCFZ2FUYYR","short_pith_number":"pith:R2NF2DKD","schema_version":"1.0","canonical_sha256":"8e9a5d0d434a691f39dac08b9d1698c4688a5325a7df985003e843b782cbc759","source":{"kind":"arxiv","id":"1901.06523","version":7},"attestation_state":"computed","paper":{"title":"Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Tao Luo, Yanyang Xiao, Yaoyu Zhang, Zheng Ma, Zhi-Qin John Xu","submitted_at":"2019-01-19T13:37:39Z","abstract_excerpt":"We study the training process of Deep Neural Networks (DNNs) from the Fourier analysis perspective. We demonstrate a very universal Frequency Principle (F-Principle) -- DNNs often fit target functions from low to high frequencies -- on high-dimensional benchmark datasets such as MNIST/CIFAR10 and deep neural networks such as VGG16. This F-Principle of DNNs is opposite to the behavior of most conventional iterative numerical schemes (e.g., Jacobi method), which exhibit faster convergence for higher frequencies for various scientific computing problems. With a simple theory, we illustrate that t"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1901.06523","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-19T13:37:39Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"3a001c56b4f0b811369a91460bcd45a110c995dfeb698c3926b6bf59acd49f47","abstract_canon_sha256":"1731f1124cd0d607dbff57ad24cfbefa3c794007836a02f953db15c9f26de404"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:21:34.692425Z","signature_b64":"Td3CExasg57Ee43CHnKbtbIjfeg+kEAAmvzSpHghUY5E2II8h4hoXT5k0lLcFJfQrypPJCFKumMUWOxw5knoBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e9a5d0d434a691f39dac08b9d1698c4688a5325a7df985003e843b782cbc759","last_reissued_at":"2026-07-05T08:21:34.692015Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:21:34.692015Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Tao Luo, Yanyang Xiao, Yaoyu Zhang, Zheng Ma, Zhi-Qin John Xu","submitted_at":"2019-01-19T13:37:39Z","abstract_excerpt":"We study the training process of Deep Neural Networks (DNNs) from the Fourier analysis perspective. We demonstrate a very universal Frequency Principle (F-Principle) -- DNNs often fit target functions from low to high frequencies -- on high-dimensional benchmark datasets such as MNIST/CIFAR10 and deep neural networks such as VGG16. This F-Principle of DNNs is opposite to the behavior of most conventional iterative numerical schemes (e.g., Jacobi method), which exhibit faster convergence for higher frequencies for various scientific computing problems. With a simple theory, we illustrate that t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.06523","kind":"arxiv","version":7},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1901.06523/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1901.06523","created_at":"2026-07-05T08:21:34.692075+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.06523v7","created_at":"2026-07-05T08:21:34.692075+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.06523","created_at":"2026-07-05T08:21:34.692075+00:00"},{"alias_kind":"pith_short_12","alias_value":"R2NF2DKDJJUR","created_at":"2026-07-05T08:21:34.692075+00:00"},{"alias_kind":"pith_short_16","alias_value":"R2NF2DKDJJUR6OO2","created_at":"2026-07-05T08:21:34.692075+00:00"},{"alias_kind":"pith_short_8","alias_value":"R2NF2DKD","created_at":"2026-07-05T08:21:34.692075+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":25,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06675","citing_title":"Spectral Born machines: classically trainable quantum generative models for discrete data","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26763","citing_title":"Calibrated Harmonic Overlaid Implicit Neural Representations for Multi-Dimensional Data","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03936","citing_title":"Correcting Neural Operator Spectral Bias via Diffusion Posterior Sampling with Sparse Observations","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02417","citing_title":"Dynamic Spectral Denoising with Global-Context Attention for Multi-Behavior Recommendation","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01622","citing_title":"An alternating learning-based collocation method for solving inverse elliptic problems","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24278","citing_title":"Fourier Feature Pyramids for Physics-Informed Neural Networks","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25057","citing_title":"Random Neural Network Expressivity for Non-Linear Partial Differential Equations","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26382","citing_title":"Detail Consistent Stage-Wise Distillation for Efficient 3D MRI Segmentation","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30375","citing_title":"Full-field prediction for engineering-scale three-dimensional aircraft with multigrid-hierarchical learning","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"1906.09235","citing_title":"Theory of the Frequency Principle for General Deep Neural Networks","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2407.12173","citing_title":"Beta Sampling is All You Need: Efficient Image Generation Strategy for Diffusion Models using Stepwise Spectral Analysis","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2410.04096","citing_title":"Sinc Kolmogorov-Arnold network and its application for solving PDEs with singularities","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2509.24814","citing_title":"A Greedy PDE Router for Blending Neural Operators and Classical Methods","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2509.23249","citing_title":"Deep Learning for Subspace Regression","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2510.25781","citing_title":"A Practitioner's Guide to Kolmogorov-Arnold Networks","ref_index":257,"is_internal_anchor":false},{"citing_arxiv_id":"2507.02592","citing_title":"WebSailor: Navigating Super-human Reasoning for Web Agent","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2603.24654","citing_title":"Spectral methods: crucial for machine learning, natural for quantum computers?","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08935","citing_title":"PnP-Corrector: A Universal Correction Framework for Coupled Spatiotemporal Forecasting","ref_index":111,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02751","citing_title":"Understanding Latent Diffusability via Fisher Geometry","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08935","citing_title":"PnP-Corrector: A Universal Correction Framework for Coupled Spatiotemporal Forecasting","ref_index":111,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10258","citing_title":"Parity Supervision as a Driver of Generalization in Quantum Generative Modeling","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24745","citing_title":"Conflict-Aware Harmonized Rotational Gradient for Multiscale Kinetic Regimes","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18686","citing_title":"Neural Spectral Bias and Conformal Correlators I: Introduction and Applications","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18673","citing_title":"Neural Networks Reveal a Universal Bias in Conformal Correlators","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11829","citing_title":"Learning on the Temporal Tangent Bundle for Physics-Informed Neural Networks","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR","json":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR.json","graph_json":"https://pith.science/api/pith-number/R2NF2DKDJJUR6OO2YCFZ2FUYYR/graph.json","events_json":"https://pith.science/api/pith-number/R2NF2DKDJJUR6OO2YCFZ2FUYYR/events.json","paper":"https://pith.science/paper/R2NF2DKD"},"agent_actions":{"view_html":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR","download_json":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR.json","view_paper":"https://pith.science/paper/R2NF2DKD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.06523&json=true","fetch_graph":"https://pith.science/api/pith-number/R2NF2DKDJJUR6OO2YCFZ2FUYYR/graph.json","fetch_events":"https://pith.science/api/pith-number/R2NF2DKDJJUR6OO2YCFZ2FUYYR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR/action/storage_attestation","attest_author":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR/action/author_attestation","sign_citation":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR/action/citation_signature","submit_replication":"https://pith.science/pith/R2NF2DKDJJUR6OO2YCFZ2FUYYR/action/replication_record"}},"created_at":"2026-07-05T08:21:34.692075+00:00","updated_at":"2026-07-05T08:21:34.692075+00:00"}